A Comparative Study Of User Comment Behavior On Different Social Platforms

In modern society,the connection between social media and human daily life is getting closer.On the one hand,hot events in various fields are first launched on social platforms,and social media has gradually become the main venue for public participation in social public affairs,attracting the public to express opinions and express emotions.On the other hand,the adverse phenomena such as extreme emotions that accompany them directly affect personal life,platform atmosphere and social stability.In order to resolve extreme emotions and social contradictions,it is of great practical significance to study social media user behavior.As big data research provides us with a way to explore user behavior patterns,researchers mostly use a specific social media as a data source in the research of online public opinion.However,the vigorous development of different types of social media has made multi-terminals and cross-platforms a daily use behavior of the public.The rise of short video platforms has brought a test to the traditional social media research represented by Weibo.Multi-source heterogeneous public opinion Research using a single social media as a data source poses challenges.Existing research still lacks consideration of the differences between different social platforms and the structure of user groups.Based on this,this study takes food safety issues as examples,selects Weibo,Douyin and bilibili as representatives of social platforms,analyzes the differences of user reviews of the three platforms,and tries to explore whether the differences are related to the platform.Using the crawler program to obtain comment data,under the guidance of the framework theory,the content of the comment text is analyzed,and the Bayesian classification model is used to classify the sentiment of the review,and to determine the relationship between the review frame and the emotional tendency.(1)Users of the three platforms Weibo,Douyin and bilibili have significant differences in the types of comments,comment cycles and interactive behaviors.Among them,Douyin’s comments are the most subjective,and the shortest space is used to express personal opinions and emotions.The commentary of Douyin is characterized by fluctuations that increase and decrease over time.(2)In terms of the relationship between the comment framework and emotional tendencies,the comments on the Weibo and B-stations using the Responsibility Attribution Framework are mostly positive emotions,which are more logical in language organization and higher in rationality.On the contrary,Douyin ’s comments using the Responsibility Attribution Framework tend to have negative emotional tendencies.The language logic is single but the rhetoric is intense.Exclamation marks and question marks are often superimposed to strengthen negative emotions.(3)In the face of negative news such as food safety issues,although negative emotions are still the norm in social media reviews,rational opinion expression has been recognized by more and more users,indicating that rationality is gradually returning.Based on the research findings,three suggestions are proposed: one is to focus on the multi-source and heterogeneous characteristics of public opinion information in public opinion research,and establish an effective information dissemination effect evaluation system;the second is to use social media to increase the use of social media with the public when disseminating public issues.The third is to build a differentiated operation strategy based on platform differences in new media operations.